在线阅读 --自然科学版 2020年3期《基于峰度ICA和本征图像滤波的超高斯信号去噪方法》
基于峰度ICA和本征图像滤波的超高斯信号去噪方法--[在线阅读]
孙婷婷1, 崔少华2
1. 淮北职业技术学院 计算机系, 安徽 淮北 235000;
2. 淮北师范大学 物理与电子信息学院, 安徽 淮北 235000
起止页码: 209--214页
DOI: 10.13763/j.cnki.jhebnu.nse.2020.03.004
摘要
针对超高斯信号的分析领域中,传统独立分量分析法(ICA)提取的独立分量信噪比低、耗时长的问题,提出了基于峰度的ICA算法,以混合超高斯信号的统计特性为基础,采用峰度作为超高斯性的唯一判断依据,首先经过本征图像滤波压缩信源空间,然后经过球化消除信源的二阶相关性,最后通过迭代判断最大峰度,对应输出一个独立分量.超高斯地震信号实验表明,所提算法有效可行,与传统ICA、基于负熵的ICA相比,迭代次数更少,输出信噪比更高.

De-noising Method for Super Gaussian Signal Based on Kurtosis ICA and Intrinsic Image Filtering
SUN Tingting1, CUI Shaohua2
1. Department of Computer Science, Huaibei Vocational and Technical College, Anhui Huaibei 235000, China;
2. College of Physics and Electronic Information, Huaibei Normal University, Anhui Huaibei 235000, China
Abstract:
In the field of super Gaussian signal analysis,the signal-to-noise ratio of independent components extracted by traditional ICA is low and the process is time-consuming.To solve these problems,we propose an ICA algorithm based on kurtosis.Based on the statistical characteristics of the mixed super Gaussian signal,the proposed method uses kurtosis as the only judgment basis of super Gaussian.Firstly,the source space is compressed by intrinsic image filtering,then the second-order correlation of the source is eliminated by spheroidization,finally,the maximum kurtosis is determined by iteration,which output an independent component correspondingly.The experiment of super Gaussian seismic signal demonstrates that the proposed method is effective and feasible.Compared with traditional ICA and ICA based on negative entropy,it has fewer iterations and higher output SNR.

收稿日期: 2020-01-03
基金项目: 安徽省质量工程项目(2015zy094);安徽省大学生创客实验室项目(2016ckjh182)

参考文献:
[1]王家映.地球物理反演理论[M].第2版.北京:高等教育出版社,2002:139-140. WANG Jiaying.Geophysical Inversion Theory[M].2nd ed.Beijing:Higher Education Press,2002:139-140.
[2]张华,陈小宏,杨海燕.地震信号去噪的最优小波基选取方法[J].石油地球物理勘探,2011,46(1):70-75,164,170-171.doi:10.13810/j.cnki.issn.1000-7210.2011.01.020 ZHANG Hua,CHEN Xiaohong,YANG Haiyan.The Optimal Wavelet Base Selection Method for Seismic Signal Denoising[J].Petroleum Geophysical Exploration,2011,46(1):70-75,164,170-171.
[3]猷凤.齿轮箱故障诊断的小波包-ICA分析方法[D].南京:南京理工大学,2015. YOU Feng.Wavelet Packet ICA Analysis Method for Gearbox Fault Diagnosis[D].Nanjing:Nanjing University of Technology,2015
[4]崔少华,单巍,赵庆平.基于负熵的快速不动点ICA算法在去噪中的应用[J].宿州学院学报,2017,32(1):98-100.doi:10.3969/j.issn.1673-2006.2017.01.026 CUI Shaohua,SHAN Wei,ZHAO Qingping.Application of Fast Fixed Point ICA Algorithm Based on Negative Entropy in Denoising[J].Journal of Suzhou University,2017,32(1):98-100.
[5]高立坤,刘东启,陈志坚.基于相关性评估与Fastica的实时心电信号提取算法[J].传感器与微系统,2018,37(8):112-115.doi:10.13873/j.1000-9787(2018)08-0112-04 GAO Likun,LIU Dongqi,CHEN Zhijian.Real Time ECG Signal Extraction Algorithm Based on Correlation Evaluation and Fastica[J].Sensors and Microsystems,2018,37(8):112-115.
[6]FAVILLA R,ZUCCALA V C,COPPINI G.Heart Rate and Heart Rate Variability from Single-channel Video and ICA Integration of Multiple Signals[J].IEEE Journal of Biomedical and Health Informatics,2019,23(6):2398-2408.doi:10.1109/JBHI.2018.2880097
[7]GLASSER M F,COALSON T S,BIJSTERBOSCH J D,et al.Classification of Temporal ICA Components for Separating Global Noise from Fmri Data:Reply to Power[J].Neuroimage,2019,197:435-438.doi:10.1016/j.neuroimage.2019.04.046
[8]陈军林,彭润民,闫岩,等.基于独立成分分析和形态特征的干涸水系提取[J].国土资源遥感,2018,30(4):163-170.doi:10.6046/gtzyyg.2018.04.24 CHEN Junlin,PENG Runmin,YAN Yan,et al.Extraction of Dry Water System Based on Independent Component Analysis and Morphological Characteristics[J].Remote Sensing of Land and Resources,2018,30(4):163-170.
[9]武爱文.概率论与数理统计[M].上海:上海交通大学出版社,2014:161-165. WU Aiwen.Probability Theory and Mathematical Statistics[M].Shanghai:Shanghai Jiaotong University Press,2014:161-165
[10]崔少华,单巍,方振国,等.基于SVD的频域滤波去噪算法[J].河北师范大学学报(自然科学版),2018,42(2):126-130.doi:10.13763/j.cnki.jhebnu.nse.2018.02.006 CUI Shaohua,SHAN Wei,FANG Zhenguo,et al.Frequency Domain Filtering and Denoising Algorithm Based on SVD[J].Journal of Hebei Normal University(Natural Science),2018,42(2):126-130.
[11]洪斌,邓波,彭甫阳,等.基于PCA降维的云资源状态监控数据压缩技术[J].计算机科学,2016,43(8):19-25.doi:10.11896/j.issn.1002-137X.2016.08.004 HONG Bin,DENG Bo,PENG Fuyang,et al.Data Compression Technology of Cloud Resource State Monitoring Based on PCA Dimension Reduction[J].Computer Science,2016,43(8):19-25.
[12]郑升,李月,董新桐.Shearlet域深度残差CNN用于沙漠地震信号去噪[J].吉林大学学报(信息科学版),2019,37(1):1-7.doi:10.19292/j.cnki.jdxxp.2019.01.001 ZHENG Sheng,LI Yue,DONG Xintong.The Application of Depth Residual CNN in Shearlet Domain to the Denoising of Desert Seismic Signal[J].Journal of Jilin University(Information Science),2019,37(1):1-7.